In the 1920s, psychologist Henry Goddard administered intelligence tests to Europeans arriving at Ellis Island, the main immigration gateway to the United States, and concluded that the overwhelming majority of Jews, Hungarians, Italians, and Russians were “feebleminded.” Goddard also used his tests to identify feebleminded American families (who, he claimed, were largely responsible for the nation’s social problems) and suggested that the government should segregate them in isolated colonies and “take away from these people the power of procreation” (Goddard, 1913, p. 107). The United States subsequently passed laws restricting the immigration of people from Southern and Eastern Europe, and 27 states passed laws requiring the sterilization of “mental defectives.” Although Canada did not exclude immigrants to the same degree as the United States, sterilization laws were also passed in British Columbia and Alberta in the 1920s. In Alberta, an amendment in 1937 allowed for the sterilization of “mental defectives” without their consent, and this law stayed on the books until 1972. Over 2800 procedures were performed in total, including 50 in the last year of the law.

From Goddard’s day to our own, intelligence tests have been used to rationalize prejudice and discrimination against people of different races, religions, and nationalities. Although intelligence testing has achieved many notable successes, its history is marred by more than its share of fraud and disgrace (Chorover, 1980; Lewontin, Rose, & Kamin, 1984). The fact that intelligence tests have occasionally been used to further detestable ends is especially ironic because, as you are about to see, such tests were originally developed for the most noble of purposes: to help underprivileged children succeed in school.

At the end of the nineteenth century, France instituted a sweeping set of education reforms that made a primary school education available to children of every social class, and suddenly French classrooms were filled with a diverse mix of children who differed dramatically in their readiness to learn. The French government called on Alfred Binet and Theodore Simon to create a test that would allow educators to develop remedial programs for those children who lagged behind their peers (Siegler, 1992). “Before these children could be educated,” Binet (1909) wrote, “they had to be selected. How could this be done?”

Binet and Simon worried that if teachers were allowed to do the selecting, then the remedial classrooms would be filled with poor children, and that if parents were allowed to do the selecting, then the remedial classrooms would be empty. So they set out to develop an objective test that would provide an unbiased measure of a child’s ability. They began, sensibly enough, by looking for tasks that the best students in a class could perform and that the worst students could not—in other words, tasks that could distinguish the best and worst students and thus predict a future child’s success in school. The tasks they tried included solving logic problems, remembering words, copying pictures, distinguishing edible and inedible foods, making rhymes, and answering questions such as, “When anyone has offended you and asks you to excuse him, what ought you to do?” Binet and Simon settled on 30 of these tasks and assembled them into a test that they claimed could measure a child’s “natural intelligence.” What did they mean by that phrase?

We here separate natural intelligence and instruction…by disregarding, insofar as possible, the degree of instruction which the subject possesses.…We give him nothing to read, nothing to write, and submit him to no test in which he might succeed by means of rote learning. In fact, we do not even notice his inability to read if a case occurs. It is simply the level of his natural intelligence that is taken into account. (Binet, 1909)

Binet and Simon designed their test to measure a child’s aptitude for learning independent of the child’s prior educational achievement, and it was in this sense that they called theirs a test of natural intelligence. They suggested that teachers could use their test to estimate a particular child’s “mental level” simply by computing the average test score of many children in different age groups and then finding the age group whose average test score was most like that of the particular child’s. For example, a child who was 10 years old but whose score was about the same as the score of the average 8-year-old was considered to have the mental level of an 8-year-old and thus to need remedial education.

German psychologist William Stern (1914) suggested that this mental level could be thought of as a child’s mental age and that the best way to determine whether a child was developing normally was to examine the ratio of the child’s mental age to the child’s physical age. American psychologist Lewis Terman (1916) formalized this comparison by developing a statistic called the intelligence quotient or IQ score. There are two ways to compute an IQ score, each of which has a unique problem.

Each of the methods for computing IQ has a problem—but luckily, each has a different problem so they can be used in combination. Psychologists typically compute ratio IQ when testing children and compute deviation IQ when testing adults.

Most modern intelligence tests have their roots in the test developed more than a century ago by Binet and Simon. For instance, the Stanford–Binet Intelligence Scale is based on Binet and Simon’s original test and was initially updated by Lewis Terman and his colleagues at Stanford University. Probably the most widely used modern intelligence test is the Wechsler Adult Intelligence Scale (WAIS), named after its originator, psychologist David Wechsler. Like Binet and Simon’s original test, it measures intelligence by asking respondents to answer questions and solve problems. Respondents are required to see similarities and differences between ideas and objects, to draw inferences from evidence, to work out and apply rules, to remember and manipulate material, to construct shapes, to articulate the meaning of words, to recall general knowledge, to explain practical actions in everyday life, to use numbers, to attend to details, and so forth. In the spirit of Binet and Simon’s early test, none of the tests require writing words. Some sample problems from the WAIS are shown in TABLE 10.1.

These sample problems may look to you like fun and games (perhaps minus the fun part), but decades of research show that a person’s performance on tests like the WAIS predict a wide variety of important life outcomes (Deary, Batty, & Gale, 2008; Deary, Batty, Pattie, & Gale, 2008; Der, Batty, & Deary, 2009; Gottfredson & Deary, 2004; Leon et al., 2009; Richards et al., 2009; Rushton & Templer, 2009; Whalley & Deary, 2001). For example, intelligence test scores are excellent predictors of income. One study compared siblings who had significantly different IQs and found that the less intelligent sibling earned roughly half of what the more intelligent sibling earned over the course of their lifetimes (Murray, 2002) (see FIGURE 10.1).

One reason for this is that intelligent people have a variety of traits that promote economic success. For example, they are more patient, they are better at calculating risk, and they are better at predicting how other people will act and how they should respond (Burks et al., 2009). But the main reason why intelligent people earn much more money than their less intelligent counterparts (or siblings!) is that they get more education (Deary et al., 2005; Nyborg & Jensen, 2001). In fact, a person’s IQ is a better predictor of the amount of education he or she will receive than is the person’s social class (Deary, 2012; Deary et al., 2005). Intelligent people spend more time in school and perform better when they are there: The correlation between IQ and academic performance is roughly r = 0.50 across a wide range of people and situations. This continues after school ends. Intelligent people perform so much better at their jobs (Hunter & Hunter, 1984) that one pair of researchers concluded that “for hiring employees without previous experience in the job, the most valid measure of future performance and learning is general mental ability” (Schmidt & Hunter, 1998, p. 262) (see The Real World box).

Intelligent people are not just wealthier, they are healthier as well. Researchers who have followed millions of people over decades have found a strong correlation between intelligence and both health and longevity. Intelligent people are less likely to smoke, and are more likely to exercise and eat well (Batty et al., 2007; Weiser et al., 2010). Not surprisingly, they also live longer. In fact, every 15 point increase in a young person’s IQ is associated with a 24 percent decrease in his or her ultimate risk of death from a wide variety of causes, including cardiovascular disease, suicide, homicide, and accidents (Calvin et al., 2010). Health and wealth are related, of course, and some data suggest that intelligence promotes longevity by allowing people to succeed in school, which allows them to get better jobs, which allows them to earn more money, which allows them to avoid illnesses such as cardiovascular disease (Deary, Weiss, & Batty, 2011). However it happens, the bottom line is clear: Intelligence matters. In case you are now seriously considering ways to change your IQ (see the Other Voices feature at the end of this chapter), remember that intelligence is only one of many factors that predict important life outcomes. For example, you learned in the Emotion and Motivation chapter that the ability to delay gratification is a powerful predictor of academic, health, and financial outcomes (Mischel et al., 2004; Duckworth & Seligman, 2005). Recent research suggests that psychological well-being (happiness) is another very important predictor (De Neve & Oswald, 2012).